DE60311574T2 - Time monitoring of packet re-transmissions during a soft handoff - Google Patents

Abstract

The present invention relates to a method for scheduling data retransmissions, a method for use in a data retransmission scheme and a method for updating a soft buffer of a base station in a mobile communication system during a soft-handover. The present invention relates to a base station executing the controlling and updating method, a communication terminal for executing the scheduling method and to a mobile communication system comprising at least one the base station and communication terminal. To prevent erroneous combining of data packets in a packet retransmission scheme at the receiver, the present invention provides a method that may flush the soft buffer region associated to a received data packet upon its correct reception. Further, a method is provided that monitors the time elapsed since the last storing of a data packet in a buffer region of a base station to be able to trigger the flush of the buffer region upon expiry of a threshold time period.

Description

The The present invention relates to a method for scheduling of data retransmissions and a method for use in a packet retransmission protocol in a communication terminal, which Part of a mobile communication system is the communication terminal and a A plurality of base stations, wherein the communication terminal during a Soft handover with the plurality of base stations communicates. Furthermore The present invention relates to a method for updating a soft buffer of a base station, which is part of the mobile communication system is. Furthermore, the present invention relates to a base station, which executes the method for controlling data retransmissions, as well as a communication terminal, which performs the method of scheduling data retransmissions. Finally, concerns the present invention is a mobile communication system that at least a base station and at least one communication terminal.

BACKGROUND OF THE TECHNIQUE

W-CDMA (Wideband Code Division Multiple Access) is a radio interface for IMT-2000 (International Mobile Communication) standardized for use as a 3rd generation wireless mobile telecommunications system. It provides a variety of services, such as voice services and multimedia mobile communication services, in a dynamic and efficient manner. The standardization bodies in Japan, Europe, the USA and other countries have jointly organized a project called the 3 ^rd Generation Partnership Project (3GPP) to produce common radio interface specifications for W-CDMA.

The standardized European Version of IMT-2000 is generally referred to as UMTS (Universal Mobile Telecommunication System). The first version of the UMTS specification was published in 1999 (release 99). In the meantime, several improvements have been made to the standard the 3GPP in Release 4 and in Release 5 standardized and discussions about more Improvements are currently under release 6.

Of the dedicated channel (DCH) for the Downlink and the uplink as well as the downlink shared channel (DSCH) were defined in Release 99 and Release 4. In the following years the developers realized that for the provision of multimedia services - or other services in the General - one asymmetric high-speed access must be implemented. In Release 5, the High Speed Downlink Packet Access (HSDPA) presented. The new high-speed downlink shared channel (HS-DSCH) offers a high-speed access in the downlink to the user of the UMTS Radio Access Network (RAN) to the communication terminals, the in the UMTS specifications with User Equipment (UE) become.

HSDPA based on methods such as fast scheduling of Packets, adaptive modulation and the hybrid ARQ (HARQ) method, to achieve high throughput, reduce delays and to achieve peak data rates.

HYBRID ARQ PROTOCOLS

The common Method for error detection in services without real-time conditions (non-real-time services) based on ARQ (Automatic Repeat Request) protocols, the ones with the forward error correction (Forward Error Correction - FEC) are combined and referred to as hybrid ARQ. When a Error is detected by the Cyclic Redundancy Check (CRC), demands the recipient the transmitter on, additional Bits or to send a new data packet. From the different ones existing protocols, the continuous ARQs Stop-and-Wait (SAW) and Selective Repeat used in mobile communication.

A Data unit is encoded before sending. Depending on the retransmitted bits can three different ARQ types are defined.

at the HARQ type I will be the received faulty data packets, also known as PDUs (Packet Data Unit), discarded and a new copy of this PDU is sent repeatedly and decoded separately. There will be no combining earlier or later Versions of this PDU performed. Using the Type II HARQ protocol, the faulty PDU, which must be sent repeatedly, not discarded, but with some retransmit bits sent from the sender to the subsequent decodes are provided. Repeatedly sent PDUs may have higher ones Encoding rates and are stored in the receiver with the Values combined. This means, that only a little redundancy at each retransmission added becomes.

Finally, the Type III HARQ protocol is almost the same packet retransmission protocol as Type II, and only differs in that each repeatedly transmitted PDU is itself decodable. This implies that the PDU is decodable without combining with previous PDUs. If some PDUs are so badly damaged that almost no information is reusable, self-decodable packets can be advantageously used.

UMTS ARCHITECTURE

The high-level R99 / 4/5 architecture of the Universal Mobile Telecommunication System (UMTS) will be released in 1 (See 3GPP TR 25.401: "UTRAN Overall Description", available at http://www.3gpp.org.) The network elements are functional in the core network (CN) 101 , the UMTS Terrestrial Radio Access Network (UTRAN) 102 and the User Equipment (UE) 103 grouped. The UTRAN 102 is responsible for controlling all radio-related functionality while the CN 101 is responsible for routing calls and data connections to external networks. The connections between these network elements are defined by open interfaces (Iu, Uu). It should be noted that the UMTS system is modular and therefore it is possible to have multiple network elements of the same type.

2 illustrates the current UTRAN architecture. A number of RNCs (Radio Network Controllers) 201 . 202 are with the CN 101 connected. Every RNC 201 . 202 controls one or more base stations (Node Bs) 203 . 204 . 205 . 206 which in turn communicate with the UEs. An RNC controlling multiple base stations is referred to as controlling RNC (C-RNC) for these base stations. A number of controlled base stations connected to their C-RNC are called Radio Network Subsystem (RNS) 207 . 208 designated. For each connection between the User Equipment and the UTRAN, an RNA is the Serving RNA (S-RNA). It receives the so-called Iu connection with the core network (CN) 101 upright. If necessary, the drift supports RNA (D-RNA) 302 Serving RNA (S-RNA) 301 by providing radio resources, as in 3 shown. The corresponding RNCs are referred to as Serving RNC (S-RNC) and Drift RNC (D-RNC). It is also possible and often the case that C-RNC and D-RNC are identical and therefore the abbreviations S-RNC or RNC are used.

REFINED UTRAN ARCHITECTURE

Currently, the feasibility study for the further development of the UTRAN architecture is performed by the current R99 / 4/5 UMTS architecture (see 3GPP TSG RAN WG3: "Feasibility Study on the Evolution of the UTRAN Architecture", available at http://www.global.com/) Two general proposals for the advanced architecture (see 3GPP TSG RAN WG3, Meeting # 36, Proposed Architecture on UTRAN Evolution, Tdoc R3-030678, and Further Clarifications on Presented Evolved Architecture, Tdoc R3). 030688, available at http://www.3gpp.org) have emerged and the proposal entitled "Further Clarifications on the Presented Evolved Architecture" will be referred to below 4 discussed.

The RNG (Radio Network Gateway) 401 is used to work with the traditional RAN and acts as a Mobility Anchor Point, meaning that if an RNG 401 for the connection is selected, it is maintained for the duration of the call. This includes functions both at the control plane and at the user plane.

• – Iu-Signalisierungs-Gateway, das heißt, Ankerpunkt für die RANAP-(Radio Access Network Application Port)Verbindung, – RANAP-Verbindungsbeendigung, umfassend: – Aufbau und Freigabe der Signalisierungsverbindungen – Benachteiligen verbindungsloser Nachrichten – Verarbeiten von verbindungslosen RANAP-Nachrichten, – Übertragen der Paging-Nachricht im Idle Mode und Connected Mode (im verbindungsfreien und verbundenen Modus) zu dem/den relevanten NodeB+(s),
• – Das RNG übernimmt die Rolle des CNs bei dem Wechsel (relocation) zwischen den NodeB+s,
• – Steuern der Benutzerebene (control plane) und
• – Iur-Signalisierungs-Gateway zwischen NodeB+ 402–405 und R99/4/5-RNC.

At the control level, the RNG acts 401 as a signaling gateway between the evolved RAN and the CN as well as the enhanced RAN and the R99 / 4/5 UTRAN. It has the following main functions:

• - Iu signaling gateway, that is, anchor point for the RANAP (Radio Access Network Application Port) connection, - RANAP connection termination, comprising: - establishing and releasing the signaling links - Disadvantageous connectionless messages - processing connectionless RANAP messages, - Transmitting the paging message in idle mode and connected mode (in connectionless and connected mode) to the relevant NodeB + (s),
• The RNG assumes the role of the CN in the relocation between the NodeB + s,
• - controlling the user plane (control plane) and
• - Iur signaling gateway between NodeB + 402 - 405 and R99 / 4/5-RNC.

• – Schalten des Verkehrs auf der Benutzerebene während des Wechsels
• – Übertragen von GTP-(GPRS Tunneling Protocol auf der Iu-Schnittstelle)Paketen zwischen NodeB+ und SGSN (Serving GPRS Support Node, ein Element des CNs) und
• – Iur-Zusammenarbeit für die Benutzerebene (control plane).

The RNG is also the user level access point from the CN or the conventional RAN to the evolved RAN. It has the following user-level functions:

• - switching traffic at the user level during the change
• - Transmission of GTP (GPRS Tunneling Protocol on the Iu Interface) packets between NodeB + and SGSN (Serving GPRS Support Node, an element of the CNs) and
• - Iur cooperation for the user level (control plane).

The NodeB + element 402 - 405 completed all of the RAN radio protocols (Layer 1 - Physical Layer, Layer 2 - MAC (Medium Access Control) and RLC (Radio Link Control) sublayers, and Layer 3 - Radio Resource Control). The control plane functions of NodeB + s 402 - 405 include all of the functions related to the control of the terminals in the connected mode within the enhanced RAN.

• – Steuern des UEs,
• – RANAP-Verbindungsbeendigung, – Verarbeiten von verbindungsorientierten RANAP-Protokollnachrichten,
• – Steuern/Beenden der RRC-(Radio Resource Control)Verbindung und
• – Steuern der Initialisierung der relevanten Benutzerebenenverbindungen.

The main functions are:

• - controlling the UEs,
• RANAP connection termination, processing of connection-oriented RANAP protocol messages,
• - controlling / terminating the RRC (Radio Resource Control) connection and
• Control the initialization of the relevant user plane connections.

The UE context is removed from the (Serving) NodeB + when the RRC connection is terminated, or when the functionality is allocated to another NodeB + (Serving NodeB + switch). The control plane functions also include all of the functions for controlling and configuring the resources of the NodeB + cells 402 - 405 and the allocation of the dedicated resources upon request from the control plane part of the Serving NodeB +. The "+" in the term "NodeB +" represents the extended functionality of the base station compared to the R99 / 4/5 specifications.

The user-level functions of the NodeB + 402 - 405 include the protocol functions of PDCP (Packet Data Convergence Protocol), RLC (Radio Link Control) and MAC (Medium Access Control) as well as Macro Diversity Combining.

ENHANCED UPLINK DEDICATED CHANNEL (E-DCH)

improvements the uplink for Dedicated Transport Channels (DTCH) currently being investigated by the 3GPP Technical Specification Group RAN (See 3GPP TR 25.896: "Feasibility Study for Enhanced Uplink for UTRA FDD (Release 6) ", available at http://www.3gpp.org). Because the use of IP-based Services is becoming increasingly important, there is an increased demand for Improving the reach and throughput of the RAN as well as the Reduction of the delay the dedicated uplink transport channels. Streaming, interactive and background services benefit from this enhanced uplink.

A Improvement is the use of Adaptive Modulation and Coding Schemes (AMC) in conjunction with Node B controlled scheduling (Node B controlled scheduling), and consequently an improvement in the Uu interface. As mentioned in the previous section, is the control of the maximum data rates in the uplink in the existing R99 / R4 / R5 system in the RNC. By laying the Schedulers in the Node B can reduce the latency due to the Signaling on the interface between the RNC and the node B is introduced, can be reduced, causing the scheduler in the Location is faster on temporal changes of the uplink load in the Uplink (uplink load) to respond. This will set the total latency reduced in the communication of the UE with the RAN. For this Reason can be due to the controlled by the Node B planning the interference in Uplink better controlled and the variance of the noise increase (noise rise variance) by quickly allocating higher data rates when reduces the load in the uplink, or by limiting the Data rates in the uplink, if the load in the uplink increases, better smoothed become. The range and the cell throughput can be improved by better control the interference in the uplink can be improved.

A further procedure, which is to reduce the delay in Uplink can be considered, is introducing a shorter transmission time interval (transmission time Interval - TTI) Length for the E-DCH in comparison to other transport channels. A TTI length of 2 ms is currently for the use on the E-DCH, while generally a TTI length of 5 ms on the other channels is used. The hybrid ARQ protocol, which is one of the key technologies at HSDPA is also for Considered Enhanced Uplink Dedicated Channel (E-DCH). The hybrid ARQ protocol between the Node B and the UE allows fast retransmissions of incorrectly received data units, reducing the number of RLC (Radio Link Control) retransmissions and the associated delays be reduced. This may be the one perceived by the user Quality of service (Quality of Service - QoS) improve.

Around to support the improvements described above will introduced a new MAC sublayer, hereafter referred to as MAC-eu. The units of this new sub-layer, which is described in more detail in the following sections can be described be arranged in the UE and the Node B. On the UE side, the MAC-eu leads the new task of multiplexing upper layer data (for example, the data of the MAC-d sublayer) into the new extended transport channels and the Operate the HARQ protocol transmission units.

E-DCH MAC ARCHITECTURE IN THE UE

5 exemplarily shows the entire E-DCH MAC architecture on the UE side. A new MAC functional unit, the MAC-eu 503 , is added to the MAC architecture of Rel / 99/4/5. The functional unit MAC-eu 503 becomes more detailed in 6 (See 3GPP TSG RAN WG 1, Meeting # 31: "HARQ Structure", Tdoc R1-030247, available at http://www.3gpp.org).

It M gives different data streams (MAC-d), transport the data packets that pass from the UE to the Node B. send. These data streams can different qualities of service (Quality of Service - QoS) have, for example, delay and error requirements, and can require a different configuration of HARQ instances. For this Reason can the data packets are stored in different priority queues. The Row of HARQ transmit and receive units, each in the UE and Node B is called a HARQ process. The scheduler takes into account in the assignment of HARQ processes to different priority queues Quality of service parameters. The MAC eu unit receives over the Layer 1 signaling Coordination information from Node B (Network side).

E-DCH MAC ARCHITECTURE IN THE UTRAN

at a soft handover process, the MAC-eu units in the E-DCH-MAC architecture on the UTRAN side Node B (MAC-eub) and S-RNC (MAC-eur). The scheduler in the Node B chooses the active users and leads controlling the rate by determining and signaling a commanded one Rate, a suggested rate or a TFC (Transport Format Combination - transport format combination) Threshold by, the or the active users (UEs) to a subset of the set of transport format combinations (Transport-Format Combination Set - TFCS) limited, which are allowed to send.

Each MAC eu unit corresponds to one user (UE). In 7 the Node B MAC eu architecture is presented in more detail. It should be noted that each HARQ receiver unit is assigned a particular soft buffer memory size or range for combining the bits of the packets of pending retransmissions. When a packet is successfully received, it is forwarded to the reorder buffer, which is responsible for the ordered transmission to the higher layer. According to the illustrated implementation, the reorder buffer is in soft-handover in the S-RNC. In 8th FIG. 3 illustrates the S-RNC MAC eu architecture that includes the reordering buffer of the corresponding user (UE). The number of reordering buffers corresponds to the number of data streams in the corresponding MAC eu unit on the UE side. Data and control information is sent from all Node Bs within the Active Set to the S-RNC during the soft handover.

It should be noted that the required soft buffer size of the used HARQ protocol is needed, for example a HARQ protocol that uses incremental redundancy (IR), more Soft Buffer as a protocol with Chase Combining (CC).

E-DCH SIGNALING

The control signaling associated with the E-DCH necessary for the functioning required by a particular protocol consists of uplink signaling and the downlink signaling. The signaling depends on considered Improvements in the uplink from.

Around the controlled by the Node B planning (for example, by Node B controlled time and rate scheduling), the UE needs some request messages in the uplink to send data to the Node B. The request message can provide status information of a UE, such as buffer status, Performance status and channel quality estimation included. Based on this information, the Node B can boost noise estimate and the UE "schedulen". With an assignment message, the downlink from the Node B is sent to the UE, the node points B the UE the TFCS with the maximum data rate and time intervals to which the UE is allowed to send.

in the Uplink requires the UE to display rate indicator message information signal the Node B, which to correctly decode the sent packets is required, for example, the transport block size (Transport Block Size - TBS), the modulation coding scheme (MCS) stage and so on. Furthermore in the case where HARQ is used, the UE must be HARQ-related Control information (for example, the HARQ process number, the HARQ Sequence Number, New as New Data Indicator (NDI) for UMTS Rel. 5, the redundancy version (RV), the rate matching parameters and so on).

After receiving and decoding the transmitted packets on the Enhanced Uplink Dedicated Channel (E-DCH), the Node B must inform the UE if the transmission was successful, by sending an ACK / NACK (confirmation / negative acknowledgment) in the downlink.

MOBILITY MANAGEMENT WITHIN THE R99 / 4/5-UTRAN

In this section will be first some often used terms briefly defined as well as some with mobility management Related Procedures (see 3GPP TR 21.905: "Vocabulary for 3GPP Specifications ", available at http://www.3gpp.org).

A Radio communication can be a logical connection between a single UE and a single UTRAN access point. Your physical Implementation includes radio bearer transmissions.

One Handover can be considered as changing a user connection from one Radio Bearer be defined to another. A "hard handover" becomes a new radio connection built up. In contrast, in a "soft handover" (SHO) radio links established and aborted, so that the UE always at least one Maintains radio communication with the UTRAN. The soft handover is for networks, using the CDMA (Code Division Multiple Access) technology, specific. Execution The handover is usually controlled by the S-RNC in a mobile network.

The "Active Set "includes a Set of radio links simultaneously into a specific communication service between the UE and the radio network, for example during the Soft Handover, where the UE's Active Set is all Radio links to the Node Bs of the RAN serving the UE.

method to update the Active Set (active set update) can be used be to the Active Set of communication between the UE and to change the UTRAN. The method can have three functions: adding Radio communications, removal of radio communications, as well as the combined Add and removing radio links. The maximum number of simultaneous Radio communications is generally set at four. New radio connections can added to the Active Set when the pilot signal strengths the respective base stations with respect to a certain threshold on the pilot signal of the strongest Exceed element in the Active Set. A radio connection can be removed from the Active Set if the pilot signal strength the respective base station with respect to a certain threshold in the strongest Element of the Active Set exceeds.

Of the Threshold for The addition a radio link may typically be higher than that for the removal a radio connection are selected. As a result, the results of adding and removing form one Hysteresis in terms of pilot signal strengths.

The Pilot signal measurements are sent to the network (S-RNC) from the UE the RRC signaling reported. Before sending the measurement results usually becomes a certain Filtering performed, to determine the average of fast fading. The typical Filter duration is approximately 200 ms and carries to the handover delay at (see 3GPP TS 25.133: "Requirements for Radio Resource Management (FDD), available at http://www.3gpp.org). Based on the measurement results, the S-RNC can decide to execute a the functions of the Active Set update process (Add / Remove a Node B to / from the current Active Set).

E-DCH OPERATION WHILE THE SOFT HANDOVER

The Support Soft handover is desirable in order to gain macro-diversity. For HSDPA, for example, there is no soft handover for the HS-DSCH (High Speed downlink shared channel) Transport channel supported. The Applying the soft handover causes the problem of assigning of scheduling responsibilities to all Node Bs of the Active Set and it would be an extremely tight temporal Coordination will be required to make the planning decision all the elements of the Active Set, even if the distribution of the scheduling feature clarified is. Only one Node B transmits on the HS-DSCH to an UE and consequently No gain is achieved through macrodiversity. If the UE is in the Soft handover area for dedicated Channels enters, Node B allowed to send on the HS-DSCH must be determined. The selection of Serving Node B can be either from the UE side or from the Network side (by RNC) performed become.

at In the FCS (Fast Cell Selection) method for HS-DSCH, the UE selects the cell which best for the sending of data is suitable. The UE periodically checks the channel conditions in the cells within the Active Set to check if there is a cell with better ones Channel conditions than the current radio cell (serving cell).

If the soft handover for the uplink is not supported, a serving node B must be selected. One problem that may occur is the incorrect selection of Serving Node B. For this reason, there may be a cell in the Active Set Thus, the transmission of data to a cell controlled by the current serving Node B may fail, while the transmission controlled by the other Node Bs may fail Cells is successful. The accuracy of the selection depends on several factors, such as the signaling delay and the filtering of the results of the measurements and so on.

In summary can be said to support the soft handover process for the E-DCH because of the benefits of macrodiversity, and thereby of benefit, that possible transmission errors due to a wrong selection of the best Serving Node B for the uplink can be avoided.

SOFT-HANDOVER PROCESS WITHOUT SOFT BUFFER SYNCHRONIZATION

In 9 A flow chart for the Node B soft handover process without soft buffer synchronization is shown assuming the R99 / R4 / R5 architecture. The figure represents the operation of any Node B from the Active Set.

Each Node B within the Active Set monitors in step 901 E-DPDCH (Enhanced Dedicated Physical Data Channel) for receiving uplink traffic. When in step 903 a packet within a transmission time interval (TTI) (see step 902 ), the Node B must decide whether the packet is the initial transmission or a retransmission of a previously transmitted data packet. The decision is based on the associated uplink control signaling, for example the New Data Indicator (NDI). If the received packet is a retransmission, Node B must decode the received data packet in step 905 combine with the previous programs stored in the soft buffer. If the received packet is an outbound transmission, the Node B stores (see step 906 ) the received packet in the corresponding soft buffer (possible previous transmissions stored in this soft buffer are overwritten) and can attempt to decode the packet immediately upon its receipt.

Checking whether the decoding was successful or not (see step 907 ), is performed by evaluating the CRC (Cyclic Redundancy Check) checksum. If the packet was successfully decoded, Node B routes it in step 908 to higher layers and sends it over the interface Iub / Iur to the S-RNC. If the decoding was unsuccessful, in step 909 the soft information is stored in the soft buffer.

As previously described, the soft handover process sets one additional Profit through macrodiversity ready, but to some extent complicates the system design. By doing E-DCH, for example, there is a single sending protocol unit and multiple receiving protocol entities for the soft handover process, whereas it is for a non-soft handover only one send and one receiving protocol unit.

SET UP A RADIO BEARER

In front The radio bearer (RB radio bearer) can be set up when a transmission begins and all Layers should be configured accordingly (see 3GPP TS25.331 "Radio Resource Control (RRC) protocol specification ", available at http://www.3gpp.org). The procedure for setting up a radio bearer may be appropriate vary the relationship between the radio bearer and a dedicated transport channel. Dependent on Quality of Service (Quality of Service - QoS) parameters There may be a permanently assigned dedicated channel with is associated with the RB or not.

SET UP A RADIO BEARER WITH ACTIVATION OF THE DEDICATED PHYSICAL CHANNEL

In the UMTS, the in 10 shown procedure when a new physical channel for the Radio Bearer must be created. A Radio Bearer Establishment may be initiated when an RB Establish Request Primitive is received by the Service Access Point of the higher layer on the RRC layer network side. These primitives may include a carrier reference and quality of service parameters. On the basis of these quality of service parameters, the parameters of layer 1 and layer 2 can be selected by the RRC layer on the network side.

Processing in the physical layer on the network side can be started with the CPHY-RL-Setup-Request-Primitive, which is issued to all appropriate Node Bs. If one of the intended recipients is unable to provide the service, this can be indicated in the confirmation message / confirmation (s) messages. After the establishment of the layer 1 including the beginning of the transmission / reception in the Node B, the NW RRC send a RADIO BEARER SETUP message to its peer entity (confirmed or negatively acknowledged sending is optional for NW). This message may include Layer 1, MAC, and RLC parameters. After receiving the message, the UE RRC configures layer 1 and MAC.

If the synchronization of the layer 1 is displayed, the UE can a RADIO-BEARER SETUP COMPLETE message in the acknowledged mode (confirmed Mode) send to the network. The NW RRC can handle the MAC and RLC layers on the network side.

To receiving the confirmation for RADIO BEARER SETUP COMPLETE, the RRC can create a new RLC instance using to the new Radio Bearer. The applicable procedure for the Establishment of the RLC may be by the RLC transmission mode dependent be. The RLC connection can either be established implicitly, or the explicit signaling can be used. Finally, can an RB establishment indication primitive is sent by the UE RRC and an RB Establish Confirmation Primitive may be passed through the RNC RRC be issued.

One simple HARQ method is currently only for communication between a single transmitter and a single receiver in the case of ensuring a reliable one Feedbacks possible. The feedback ensures that the sender and receiver are synchronized. By elevating the sequence number value of a window-based HARQ process or by switching the New Data Indicator (NDI) of a Stop and Wait (SAW) ARQ process in the associated HARQ control information, he knows Receiver, whether a new packet is received and whether he accordingly the Soft buffer can empty.

Thereby Ensuring that a new package does not work with one before in the receiver stored package is combined. A faulty combination packets before decoding may rarely be the case however, it may not be complete be excluded if the feedback signaling is not quite and is reliable. In this case, no correct decoding is possible.

Of the receiver can therefore a retransmission of this packet by signaling a Request NAK. The repeated transmission of this packet can be done as long done until the maximum number of retransmissions is reached is. If there are many retransmissions of a 'new' packet, that with the previous soft buffer values of one 'old' package combined, the influence of the soft values of the 'old' package may be due to the successively combining with the new package can be reduced, thereby a successful decoding of the new package is enabled. How much throughput is affected by packet retransmissions can, can from the probability of a faulty operation the packet retransmission process. It can be a conflict of goals between the effort required for the reliable one Signaling is applied, and the probability of a faulty Protocol operation exist. There may also be a procedure around the receiver about that to inform if the packet transmission was aborted by the transmitter. This For example, if the maximum number of Retransmissions is reached, or if the assigned delay attribute (or the time-to-live (TTL) value) can not be met.

Some Communication systems, such as Wideband Code Division Multiple Access (WCDMA) rely on a soft handover process. In addition to The problem is that now multiple feedbacks from each receiver are correct have to be received There is also the problem of the HARQ soft buffer between the sender and a variety of recipients to synchronize. It may be that not all Node Bs have the associated signaling can receive from the UE, the one for one correct processing of the received packet is required. Assuming that the control information was received, Node B may try to decode the packet and the soft values to buffer if successful decoding is not possible. It is likely that there is a Node B (for example the best connection) that is able to decode the packet, while others at all receive nothing.

The Sending new packets to the best Node B will continue although it is still old cached in other receivers There are packages.

In WO 02/37872 a method is presented which discloses the HARQ process during soft handover from one transmitter to multiple receivers in the uplink. The reception can not be guaranteed since the power control and hence the transmission power is normally adapted to the best connection within the Active Set. This also means that reliable feedback from all the recipients is difficult to achieve. The transmission power in the uplink must be increased for the "bad" connections to ensure well-synchronized operation, thereby significantly increasing uplink interference WO 02/37872 suggests the reliability of the HARQ protocol by adding an erase bit to increase the associated HARQ uplink control information.

One set deletion bit inform the recipient not to combine the package with the previous shipments, but clear the HARQ soft buffer of this HARQ process. This works in principle, however, there are two disadvantages. First, it is exposed to that the sender the status of the receiver knows, since he must inform him when the buffer will be emptied should. If the transmitter is due to an unreliable or missing feedback not sure regarding the status of the recipient is, the buffer should be emptied. This leads to a loss of information, if the packet has already been received and stored in the soft buffer has been. Second, this delete bit with high reliability sent together with the HARQ control information. Thereby the over-the-air signaling overhead in the uplink is increased.

The Problems with unsynchronized buffers during a multi-hand soft-handover operation Base stations acting as receivers work, were detailed described. Existing solutions based on the regular HARQ control information, such as the HARQ process and the HARQ sequence number or the NDI, on additional Signaling to clear the soft buffer and to prevent a faulty Combining.

SUMMARY THE INVENTION

It The object of the present invention is to combine incorrectly of data packets in a packet retransmission protocol (packet retransmission protocol) in the recipient. The faulty one Combining can be caused by unsynchronized soft buffers of multiple receivers become.

The Object of the present invention is achieved by the subject matter of independent claims Fulfills. Preferred embodiments The present invention is defined in the dependent claims.

If a window-based HARQ protocol as an example of a data packet retransmission protocol It should not happen that a package with the same Sequence number is received as an old packet in the soft buffer. This phenomenon is called a wrap-around problem. The HARQ window is advanced, with the soft buffer of this sequence number is not emptied. For N-channel stop-and-wait protocols the problem is similar. The same HARQ process should not be reassigned to a new package Be, except when this is displayed and the soft buffer is emptied.

The The present invention can provide a correct protocol operation with ensure multiple base stations as a data receiver, with additional signaling over the Air interface or within the network is avoided. In one First step can be any buffer after each successful decode a received data packet or a combination of a faulty one Data packets with the associated Retransmission be emptied. Additionally or as an alternative to immediately flushing the buffer when received correctly a data packet can be the time since the last save has passed in a certain buffer area, for example monitored by a timer or counter, in each base station become. This monitoring can make sure that old packets in the soft buffer are cleared before a new one Packet is received.

A Threshold time span, that is, a maximum allowable Time span after which no retransmission of a data packet may arrive more in a base station may or may not be predetermined be configured. After the expiry of this period is a associated Buffer area in the base station emptied and it can be new Data packets are received. The configuration of the threshold time can by signaling higher layers between a communication terminal, such as a UE, and a receiver, such as a base station. The initial value of the timer may correspond to the threshold time period.

As a result, can the communication terminal "know" the time to which a buffer area for a given Data packet and its associated retransmissions in the base station with which it is communicating. For this reason, it may know by what time a retransmission a specific data packet or a retransmission data packet (retransmission data packet) must be received in the base station, to benefit from the soft-combining. If the buffer in the receiver has been emptied, the communication terminal can use this knowledge in the selection the correct send parameter for use a new send of the aborted data packet.

The present invention provides a method for use in a packet retransmission protocol in a mobile communication system comprising a communication terminal and a plurality of base stations, wherein the communication ons terminal communicates with a plurality of base stations during a soft handover. The method may include the steps of receiving a data packet from the communication terminal in the plurality of base stations and checking the data integrity of the received data packet in each of the base stations. If the data integrity of the received data packet is not acknowledged by a base station, the received data packet may be stored in an area of a buffer of the respective base station, the buffer area being associated with the received data packet. The elapsed time since the data packet was stored in the associated buffer area can be monitored. It should be noted that the plurality of base stations do not have to refer to all base stations controlled by a control unit or a plurality of control units in the mobile communication network, but rather to the base stations which during the soft handover with the communication terminal communicate. In UMTS, this plurality of base stations may be referred to as the active set of the communication terminal. As a result, the plurality of base stations may be a subset of the base stations available for communication in the mobile communication network.

If the data integrity the received data packet is confirmed, the associated buffer area be emptied in the corresponding base station.

When an alternative solution the above object is the present invention of Further, a method for updating the soft buffer of a Base station in a mobile communication system ready, the Communication terminal and includes a plurality of base stations. According to this embodiment communicates with the communication terminal during a soft handover the multitude of base stations. According to the method, a data packet from the communication terminal in one Variety of base stations are received.

Furthermore can data integrity the received data packet in each of the base stations to be checked, and if the data integrity of the received data packet confirmed A buffer area can be used with the received data packet is to be emptied.

In a further embodiment of the present invention, if the data integrity of the received Data packet is not confirmed by a base station, the received data packet in an associated Area of a buffer of the respective base station are stored, and the time since the data packet was stored in the associated buffer area has elapsed, can be monitored become.

If the corresponding monitored Time span equal to one or more is a threshold time period after which a retransmission data packet can no longer be expected in the respective base station, the buffer area can be emptied. The data packet can, for example, via a dedicated channel are received.

If the data integrity a first transmission a data packet or the integrity of a retransmission data packet is not confirmed by a base station, may be a retransmission data packet in accordance with a packet retransmission protocol be requested. As a result, in another embodiment a retransmission data packet from the communication terminal in the Variety of base stations are received. Upon receipt, a Base station a data integrity check of the received retransmission data packet in each of the base stations, and if the data integrity is not confirmed by the base station if the retransmission data packet is in the buffer area, the one with a previous data packet in relation to the retransmission data packet is associated with, saved, and monitored since saving of the retransmission data packet in the associated buffer area Time can be restarted.

The Data integrity check, the in the retransmission data packet, the combining may be performed the retransmission data packet with the associated data packet, in order to obtain a combined data packet, the decoding of the combined Data packet to receive decoded data, as well as checking the Integrity of include decoded data. This generally means that Check data integrity Verify the non-corruption of the received data, the one Transmit / retransmit operation of a specific data packet correspond, for example by means of a cyclic redundancy check (Cyclic Redundancy Check - CRC).

If the data integrity a received data packet is confirmed, the associated buffer area be emptied.

Upon receiving a retransmission data packet from the communication terminal in the plurality of base stations, a data integrity check of the received retransmission data packet in each of the base stations On the other hand, if the data integrity is confirmed by a base station, the monitoring of the elapsed time since the transmission data packet was stored in the associated buffer area can be stopped. It should be further noted that in the above case, the retransmission data packet can be stored in the buffer area. The term data packet may be understood as a generic term relating to a retransmission packet or an originating message.

If the respective monitored Time span equal to one or more as a threshold time period is, monitoring can be the respective data packet also be stopped, as it is not very likely is a retransmission data packet for the data packet receive, which is associated with the buffer area. By emptying The buffer area can be assured of reuse the buffer area a new data packet not with the "old" content, that is, with a previously received data packet and its associated retransmissions, This buffer area is combined.

As As previously stated, it is desirable that the threshold period of time be more configurable Duration is.

The Signaling the duration of the threshold time period to at least one of the plurality of base stations can by radio network control signaling from a control unit in the mobile communication network become. For example, if a RAN meets the UMTS specifications may be used, the duration of the threshold period at least a base station in an information element (IE) of an NBAP (Node B Application Part) message.

Further may be signaling the duration of the threshold time period to the Communication terminal by radio resource control signaling from a control unit be accomplished in the mobile communication network. Here too can when using a RAN according to the UMTS specifications the duration of the threshold period to the communication terminal in an IE (information element) of at least one radio bearer setup message, a radio bearer reconfiguration message, a radio resource control connection setup message, a transport channel reconfiguration message, a cell update message and a handover command message.

In accordance with a packet retransmission protocol, for example HARQ the reception status of a data packet is displayed to the communication terminal become. For this reason, a message from at least one the plurality of base stations are sent to the communication terminal indicating whether at least one of the plurality of base stations is the one data integrity of the received data packet has confirmed.

One successfully received and decoded data packet can become a higher Layer forwarded for further processing. As a result, is according to an embodiment the present invention, the received data packet to a control unit the mobile communication system through at least one of the base stations which indicates the data integrity of the received data packet approved Has.

There It may happen that the capacity of the communication terminal, the for the repeated sending of a faulty data packet before emptying a buffer area assigned in the base station is insufficient the communication terminal signal the base station its assigned capacity for the retransmission data packet to increase. As a result, it receives a base station has a capacity request message from the communication terminal, the extra transmission capacity for one Request retransmission of a data packet.

The Capacity request message advantageously comprises at least one transmission priority of the communication terminal data packets to be sent, the size of the data in a transmission buffer of the communication terminal and the duration of the monitored Period of time. These parameters can advantageously be used by the base station to decide whether to increase the allocated channel capacity for the requesting communication terminal or not. Alternatively, the capacity request of the communication terminal may be appropriate a further embodiment HARQ side information, such as the sequence number, the HARQ process or the New Data Indicator (NDI) to include the package, for that's the capacity is required to identify. For identifying the package The base station may have some of the corresponding parameters of the package, such as for example, the threshold time period and the priority of the data packet, know. Likewise the communication terminal the physical channel, the transport channel and / or the logical one Channel, for it's the capacity request, identify.

In response to the capacity request message or in the case where the base station assigned the communication terminal increase capacity, a capacity allocation message is sent to the communication terminal, wherein the capacity allocation message indicates a transmission capacity assigned to the communication terminal for data transmission.

A another possibility for preventing the emptying of a buffer area, which with a specific data packet and its associated retransmission data packets may be sending a restart request message to a Base station, wherein the restart request message a Data packet displays, for that's watching since the data packet has been saved (or an associated send re-exporting packet) to restart in the associated buffer area is. A base station can receive this restart request message and the monitoring start anew. The restart request message may have control information and no or dummy payload. In a further alternative embodiment In accordance with the present invention, a base station may be at the expiration of the threshold time period an associated one Mark the buffer area of the packets as a buffer area to be emptied. If a new one with this buffer area associated packet (identified by the sequence number) received finally, she can do that Empty soft buffers unless they have some additional control information has received. Such control information may be a combination indicator. A combination indicator can be executed as a flag to be sent can if the transmitted data packet should be combined. In the latter Case, the marked buffer area must not be emptied and There is still a combination going on, although the time is already has expired. This allows a soft-combining, even if a retransmission of a data packet is delayed.

If uses a window-based packet retransmission protocol In addition, the method may further include the step of calculating the threshold time span based on the time spent on sending all data packets within a window of the packet retransmission protocol required is.

Independent of the reuse protocol used may also have the threshold time period based on the time interval between the receipt of an output data packet and the reception of a retransmission packet.

In accordance with a further embodiment may be the duration of the threshold time span calculated on the basis of at least one of the following parameters: the size of the buffer that maximum number of packet retransmissions in a data packet retransmission protocol, the processing time of the communication terminal for a feedback message, the processing time of the corresponding base station for a received Data packet and a transmission time interval.

The present invention provides above a base station in a mobile communication system ready, the one communication terminal and a plurality of base stations, wherein the communication terminal during a soft handover with the plurality of base stations communicates, and wherein the base station means for implementing the above-described Includes method.

In a further embodiment The present invention provides a method for scheduling data retransmissions in a communication terminal, which is part of a mobile communication system, wherein the Mobile communication system, the communication terminal and a A plurality of base stations, and wherein the communication terminal during a Soft handover communicates with the multitude of base stations. The method may include the step of sending a data packet of the plurality of base stations, the step of receiving at least a feedback message from at least one of the base stations, the step of evaluating the at least one feedback message, to determine whether the data integrity of the transmitted data packet by at least one of the plurality of base stations is confirmed, or whether the data integrity of the transmitted data packet is not confirmed by a base station, the step of monitoring since the transmission of the data packet or the receipt of the corresponding Feedback message elapsed time to one to the data packet associated send repeat so that they do not expire after the expiry of a Threshold period occurs after receipt of a retransmission data packet can no longer be expected in the respective base station, include.

If the data integrity the transmitted data packet is not confirmed and at a time to which the supervised Time span shorter as the threshold time period can, as already described above, a capacity request message are sent to the multitude of base stations, the additional transmission capacity for one Request retransmission of a data packet.

When a capacity allocation message indicating a transmission capacity allocated to the communication terminal for data transmission, can not be received from a base station of the plurality of base stations, or if no additional capacity is allocated to the UE in response to a capacity request message, a restart request message may be sent from the communication terminal to a base station, wherein the restart request is Message indicates a data packet for which monitoring in the respective base station is to be restarted.

The Use of a restart request message is not at the Results of a request for additional transmission capacity bound. In accordance with a further embodiment of the present invention, if the data integrity of the sent Data packet not confirmed and at a time when the monitored time period is shorter than the threshold time is a restart request message are sent to a base station, with the restart request message indicates a data packet for which monitoring to restart in the respective base station.

As later Ask in more detail can be described in the case where a retransmission packet can not be sent by the communication terminal before the associated one Buffer area in the base station is emptied, for example due to under-allocated capacity, the communication terminal on the Empty the buffer area to get a new output data packet to send. Accordingly, the method may include the step of delaying the Repeat transmissions of a data packet until the respective monitored Time lapse longer as the threshold time span is when the data integrity of the sent Data packet not confirmed and at a time when the monitored time period is shorter than is the threshold time period.

Around to initiate the sending of a retransmission data packet, if the data integrity the transmitted data packet is not confirmed, becomes a scheduler in the communication terminal informs and retransmits the sent data packet to assign. Similarly, if the data integrity for the sent Data packet confirmed the scheduler will be informed by the communication terminal, to remove the transmitted data packet from a transmission buffer of the communication terminal.

If a retransmission for the output data packet is required, the communication terminal can the retransmission data packet send to the plurality of base stations, and in turn receives at least one feedback message from at least one of the base stations. Subsequently the at least one feedback message can be evaluated, to determine if the data integrity of the retransmitted data packet being sent is confirmed by at least one of the plurality of base stations, and if the data integrity approved can, monitoring can since sending a data packet or receiving a data packet corresponding feedback message elapsed time again become. For example, if a timer is used for monitoring the timer is reset to its output threshold and restarted.

Of Further, the present invention provides a communication terminal in one Mobile communication system that provides a communication terminal and a A plurality of base stations, wherein the communication terminal during a Soft handover communicates with the multitude of base stations, and wherein the communication terminal comprises means for implementing the comprises the method steps described above.

In accordance with a further embodiment of the present invention, the Communication terminal and the base station as described above, advantageously be combined in a mobile communication system.

SHORT DESCRIPTION THE FIGURES

in the Below, the present invention will be described in more detail with reference to FIG the attached Figures and drawings described. Similar or consistent Details in the figures are provided with the same reference numbers.

1 shows the high-level architecture of UMTS,

2 shows the UTRAN architecture according to UMTS R99 / 4/5,

3 shows a drift radio subsystem and a serving radio subsystem,

4 shows the advanced UTRAN architecture,

5 the E-DCH-MAC architecture in a UE,

6 shows the MAC eu architecture in a UE,

7 shows the MAC eu architecture in a Node B,

8th shows the MAC eu architecture in an RNC,

9 shows a prior art flow diagram of the operation of a HARQ receiver,

10 shows a procedure for setting up a radio bearer according to the UMTS specifications,

11 shows a flowchart of the operation of a base station according to an embodiment of the present invention,

12 FIG. 12 is a flow chart showing the operation of a communication terminal according to an embodiment of the present invention; FIG.

13 illustrates scheduling data between a communication terminal and a base station in accordance with an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

It It should be noted that the various embodiments in the following mainly with respect to the HARQ packet retransmission protocol and the UMTS are described. Nevertheless, the principles are the present Invention, also to other data packet retransmission protocols and on mobile communication systems other than the UMTS applicable to the Providing soft handovers of communication terminals and packet retransmission mechanisms.

11 FIG. 11 shows an exemplary flowchart of the operation of a base station in the active set of a UE during a soft handover. The base station can be in step 1101 monitor the physical channels and can step in 1102 regularly check if one or more data packets are received within one transmission time interval (TTI). When a data packet is sent to the base station, it will be in step 1103 and upon reception, the base station decides in step 1104 whether the received data packet is a first transmission of a data packet (output data packet) or a retransmission with respect to an output data packet. If the received data packet is a retransmission data packet, the base station combines in step 1105 the retransmission data with the associated soft values stored in an associated soft buffer area of the base station. For example, if the first transmission of a data packet is not received correctly, that is, its data is erroneous, and can not be decoded by the base station, the retransmission data associated with that first transmission of a data packet becomes the data from the first one Transmission of the data packet combined and the combined data will be in step 1106 decoded. When a first transmission of a data packet is received, the packet may directly without a previous combining in step 1106 be decoded. In step 1106 In addition, the data integrity of the decoded data is checked.

If the data integrity is confirmed, the process goes to step 1107 above. As a first improvement compared to the in 9 The prior art packet retransmission protocol, shown in FIG. 1, becomes a buffer area associated with a data packet and its possible retransmissions, for example a HARQ soft buffer 1107 deleted or emptied immediately after correctly receiving the data packet, and not just after a new data packet is received, as is the case with conventional packet retransmission protocols (see block 906 in 9 ). Consequently, the buffer area associated with a first transmission of a data packet and its retransmissions is emptied immediately after correct decoding, thereby ensuring that no data from previously received data packets is in the buffer area when a new data packet is received associated with this buffer area.

at the transmission of data between a communication terminal and a Variety of base stations can cause the immediate emptying of a soft buffer area differ because some Node Bs temporarily have no control information can receive and the Node B receiving error-free data packets from the UE Time to continue sending.

For example, if the buffer area is emptied immediately after the data packet has been properly decoded, erroneous combining can be ruled out from the beginning. In addition, a timer may be used which is preferably started for all new packets (n = 0, 1... N) that are not successfully decoded to prevent skilful combining. After or before the soft values of a data packet, that is, an output data packet or a retransmission data packet, are stored in the soft buffer, a timer is started to indicate how long the packet has been in the buffer or how long it can still remain in the soft buffer before the associated buffer area is emptied. During normal continuous receive operation, a retransmission can be received before the timer expires and the timer for this data packet X _i is restarted. If a packet is received correctly, in addition to flushing the associated buffer area in step 1107 The timer for this data packet will be paused.

In the event that a data packet, that is, a first transmission of a data packet alone or a combination of a first transmission of a data packet with soft values of one or more retransmissions with respect to this packet, in step 1106 can not be correctly decoded, the newly received data packet is stored in an associated soft buffer area.

at the HARQ identified, for example, in each received data packet, independently of whether this is a first transfer a data packet or a retransmission data packet that HARQ process number and the NDI a retransmission of a particular Data packet as described above. If a data packet is received with a specific process number and not correct can be decoded the soft values of the data packet in an associated buffer area together with more data from packets being stored, the same HARQ process number contain.

When a timer for a data packet that can be stored in an associated buffer area is not running, that is, a first transmission of a data packet is received, the timer associated with the buffer area and the received data packet becomes step 1111 started. If the timer is already running for a received data packet associated with the buffer area, the timer can be restarted (see also step 1111 ).

Before proceeding to receive in the next transmit time interval (TTI), all timers in step 1112 be lowered. The timers may be decremented regardless of whether a data packet is received in the elapsed TTI or not.

When in step 1113 it is determined that one of the timers has expired, the particular buffer area comprising the first transmission of a data packet and possibly additional soft values of received associated retransmission packets is incremented 1114 cleared since the corresponding base station can no longer expect a retransmission data packet belonging to the erroneous data packet stored in the soft buffer area. If no timer has expired, the reception of the next data packet is expected.

One Criterion for setting the timer value, that is, for Setting a threshold time span, after that, a retransmission data packet from a base station can no longer be expected, is this time value small enough to make a mistake combining different ones To avoid packages. At the same time, the timer value should be big enough be a delete the stored packets for which there are still retransmissions queue.

For example can in HARQ protocols used in the UMTS, Release 5, HARQ broadcasts be asynchronous while ACK / NAK feedback messages can be sent synchronously. This means that the base station usually does not know exactly when a retransmission of a first transmission of a data packet through the communication terminal is sent, which makes setting the timer rather difficult becomes. If the UE can send data packets on its own, it may be that the retransmission data packet shortly after sending the Output data packet follows, with the processing time in the base station and the communication terminal considered becomes. Send retries can with a higher one Priority sent be to delays and delay variances to minimize. Furthermore, a retransmission should not wait too long, because the channel conditions can change and consequently the transmission format, for example the packet size, the modulation, the coding rate and so on, not for anymore the channel conditions may be suitable. The latter can be an adaptation the Transmission Format Combination (TFC transmission format combination) which require new channel conditions.

The maximum allowed timer value may depend on the detailed design of the packet retransmission protocol. For example, in a window-based HARQ protocol, the wrap-around problem should be avoided. For a reliable ARQ method, the ARQ window can be at least twice the size of the sender or receiver window. As soon as a data packet outside the window is received, the window is moved forward. Assuming that the sequence numbers are used to identify a particular position within the HARQ window, that the window is not advanced in larger increments, and all subsequent data packets are received correctly, the timer value may be based on the time span ne, which is required to send all the packages in the window. In the latter case, the timer would expire before a new transmission of a new data packet with the same sequence number would be initiated.

ever bigger the Window size is, or the more HARQ-SAW processes there are or the bigger the window per HARQ process and hence the soft buffer size, the higher can be the timer value. For HARQ can be the soft buffer size of a the main complexity factors be because each soft value of a retransmission data packet in a buffer area must be cached. For this Reason the window size must not be oversized be. This means that the buffer can be sized to to save as many packets as during the round trip time at continuous Send received. Dependent on from the receiver implementation the package can be implemented using bit-level or symbol-level soft values be cached. In high-level modulation (for example 16 QAM) is for caching at the symbol level requires less disk space, while Soft values at the bit level allow the highest possible flexibility for example, if certain bits need to be punctured. One Another design criterion is the bit granularity, that is, how many bits a soft value represent. As a result, there may be a trade-off between accuracy and buffer size.

In the 13 represented round-trip time can be considered the time from the first transmission 1301 a data packet from the UE to a retransmission of the same data packet 1305 be defined upon receiving a negative feedback message. This includes two times, the propagation delay t _propa plus the processing times of the UE and Node B, t _UE-process and t _{NodeB-process} . Suppose a retransmission can be sent 6 TTIs after the previous or the first transmission. To allow continuous transmission, this may correspond to a sender and receiver window size of at least 6. In a purely window-based ARQ system, the minimum total window size may be at least twice the receiver / transmitter window size, that is, 12 in the example. This ensures unambiguous packet identification and correct ARQ operation, even if some ACKs / NAKs are lost. For example, if a NAK occurs due to misinterpretation of an ACK, a new packet may not have the same sequence number, but the window would likely be moved, indicating to the receiver that this is not the expected retransmission but a new packet.

The same calculation can be performed for N-channel SAW protocols. The ARQ system described above corresponds to a 6-channel SAW protocol and a 1-bit NDI (New Data Indicator) (equal to a sequence number). In both systems, the timer can be set 12 times the TTI to make the worst case protocol work. Assuming synchronous sending, as in 13 can be calculated, the maximum timer value for the worst case. This is explained below:
The UE may send a first data packet having a process number set to 1 and an NDI equal to 0 to two Node Bs, where Node B1 receives the data packet but can not decode it and consequently the soft values in a buffer area of the soft buffer stores and sends a NAK to the UE to indicate the unsuccessful decode. Node B2 can successfully decode the data packet and sends an ACK. Consequently, the UE receives an ACK of Node B2 and the NAK of Node B1.

Of Further, it is assumed that the UE allocates the packets 2, 3, ..., 12 Node B1 and Node B2 sends, with Node B2 receiving all packets and successfully decoded. The data packets 1 to 6 can use different HARQ process numbers and an NDI set to 0. The packages 7 to 12 will be sent with the NDI set to 1.

Accepted, Node B1 may be temporary not receiving and missing the packet 7 with the process number 1 and a NDI set to 1. As a result, the first data packet remains with a process number 1 and an NDI set to 0 still in the associated Soft buffer of the Node B1 stored.

The UE sends the new data packet 13 with the process number 1 and the NDI equal to 0 received from the Node B1. Because the process number and the NDI of the new data packet of the process number and the NDI of the data packet stored in the buffer, combined Node B1 the two packets, because the new data packet as a retransmission is seen. Consequently, would Node B1 erroneously misses data packet 13 in such a situation with the data of the first data packet 1 that is in the soft buffer is stored.

This is a worst case calculation based on very specific assumptions that are rare in a real system. First, the UE need not be allocated continuously as in the previous example. The misinterpretation of the feedback signaling or the complete missing of a data packet just makes The ARQ operation may not be synchronous, meaning that the retransmission does not have to be sent after a specified time has elapsed. There may be pending retries from some other packets that have a higher priority than the outbound transmission (first transmission of a packet) and they would displace new outbound broadcasts, increasing the time to a possible wrap around. that the HARQ buffer is emptied only after a missed reception.

The used packet retransmission scheme or protocol be flexibly configurable, for example, quality of service requirements specific data streams be taken into account. This can be, for example, a certain bit error rate to be achieved or a delay request be. For example, you can the configurable parameters when using the HARQ protocol the size of the HARQ soft buffer, the Number of HARQ processes, the time to cancel a packet, the maximum number of retransmissions, the minimum intermediate TTI interval or the processing time of the UEs and / or the Node B and so on include. Setting the timer to the time at which a packet is sent from the soft buffer is, may depend on such a kind of parameters what also considered should be.

The Problem of non-synchronized Node B buffers may occur during a Soft handover occur if two or more Node Bs receive the same packet and try to to decode this. For this reason, the RNC can send the Node B over the Inform the soft handover status of each individual UE. A timer can only be started when the UE is in soft handover.

Furthermore There may be a default timer value for the Node B or the value can through higher Layers, such as the RRC (Radio Resource Control) layer, be configured. The RNC can send a message with a new one Information element, for example, named HARQ_flush_timer, signal. The message can be sent to a specific physical or transport channel, for example an E-DCH (Enhanced Dedicated Channel), set up or change. in the UMTS will use the Radio Network Control Protocol as the Node B Application Protocol (NBAP). For The packet sending the Enhanced Dedicated Uplink can have different coordination options be applied. If the Node B controls the uplink sending of the UEs, he can prioritize a particular UE whose timer is almost has expired. This allows the UE to retry the retransmission data packet before the expiration of the timer and an emptying of the soft buffer in the base stations of his Active Set to send as well as benefit from the soft-combining.

Even though Send repeats have a higher priority can, can others in the scheduling decision of the UE Parameters, such as channel quality, available transmit power, different priorities of different data streams and so on, taken into account become. If the UE can schedule some shipments on its own or additional transmission capacity It can prioritize packets whose timers are near has expired.

Thus, in another embodiment, the timer value may be known to the UE, for example, it may be predetermined. In a further embodiment of the invention, the timer value is signaled to the UE. The timer value may be signaled using the RRC signaling. This may require defining a new information element, for example called HARQ_flush_timer, for an RRC signaling message. In the 10 described procedure for setting up a carrier would not have to be changed and would integrate the new information element HARQ_flush_timer into the existing message, such as RB_setup, RB_reconfiguration, RRC_connection_setup, TrCH_reconfiguration, cell_update or handover_command. If the HARQ timer value is not signaled, it may have a default value. In addition, the granularity of the timer values may increase with higher values of the parameter HARQ_flush_timer. For example, assuming possible timer values of 5 ms, 10 ms, 15 ms, 20 ms, 30 ms, 50 ms or 100 ms, the HARQ_flush_timer can be represented by 3 bits.

Furthermore, the communication terminal may change the selection of the transmission parameters if it knows that the soft buffer has already been emptied in some or all of the Node Bs. For example, depending on how many retransmissions have been sent, there are several strategies for incremental redundancy. Some UEs send the output data packet at a low code rate to get close to the code rate needed for decoding. For further retransmission data packets, only a little redundancy is added. If it is known that a buffer area associated with a data packet and its respective retransmission data packets has been emptied in some or all of the Node Bs, the UE may revert to the first one Start transmission (of a package). The same may apply to systems that use different modulation constellations for different transmissions, such as a constellation change (see, for example, 3GPP TS 25.213: "Spreading and modulation (FDD)", available at http://www.3gpp.org).

In 12 the functionality of a HARQ transmitter is shown as an example. The UE may be ready for data transmission if the UE is synchronized, the radio bearer has been properly configured by the RRC, and so on. If the UE is allowed to send, has data in its transmit buffer, has sufficient transmit power, and so forth, it may send one or more packets within a TTI, as in block 1201 is specified. A packet is commonly referred to as a packet data unit and may be a segment of another packet, such as an IP packet, or it may also be a link of multiple packets. In step 1202 For example, the UE may select the transmission parameters such as the transport block size, the modulation coding scheme, the number of codes, the power, the constellation, and so on, and may send this side information or control information before or together with the packet described in step 1203 is sent.

There are many alternative ways that the feedback message can be generated and processed. In this example, all Node Bs send feedback and if one of the Node Bs sends an ACK (see step 1204 ), the packet is considered to be received correctly and may be taken from the sender buffer (see step 1205 ) are removed. Subsequently, in step 1206 for each node B _y within the Active Set, to see if an acknowledgment is received for the sent packet. When an ACK is received for a particular packet X _i , the corresponding timer T _{i, y} in step 1207 stopped. If no ACK is received from a base station, the process goes to block 1209 above. The steps that can be performed by all Node Bs in the Active Set (y = 0, 1 ... Y) are indicated in the figure by the different outline formats. This implies that there may be as many timers in the UE as there are Node Bs in the Active Set.

If no ACK is received for a sent packet X _i , the packet in step 1208 be assigned for a retransmission. The problem of an unsynchronized soft buffer may occur when a new data packet is sent that differs from the data packet that is still stored in the soft buffer, that is, a previously transmitted data packet. Since Node B can not know when he missed a program, the timer is started after each reception.

In step 1210 For example, the UE may start or restart the timers for each packet and Node B within the Active Set for which a negative acknowledgment NAK is received (see step 1209 ). If a Node B misses the packet (if neither an ACK nor a NAK is sent), it does not affect the timer, and the process goes to block 1211 above.

The timer may preferably be set to a multiple of the TTI and will be in step 1211 every TTI lowered.

When a timer expires in a Node B (see step 1212 ), the UE may know that the corresponding timer in that particular Node B has expired and the soft buffer for that packet has been flushed in that particular Node B (see step 1213 ). In the exemplary embodiment, the UE may send a possible retransmission, that is, in this case a first transmission of a data packet, with the output parameter settings. If nothing is coordinated within a TTI, all timers are also decreased by one TTI. The decision in the UE to restart with output send parameters may depend on other parameters or conditions, such as in how many Node Bs the buffer has been emptied, how many retransmissions have already been sent, how much the channel conditions have already changed in the meantime and so on.

It should also be noted that only a selected Node B of the Active Set can send a feedback message to the UE to indicate the receiving status (ACK, NAK) of a transmitted data packet / retransmission data packet. In the latter case, only one timer can be maintained for each sent packet. In this case, it must be mentioned that there is no guarantee that the soft buffers of the other Node Bs that are not sending a feedback message are still fully synchronized. In this case, it may be advantageous to use the invention described above or the synchronization of soft buffers by the signaling between the Node Bs of the Active Set, which in the parallel application "Base Station synchronization during Soft Handover", EP 1 507 421 A1 is described.

The following describes other operations that can be performed by the UE. These are mainly processes that prevent Node B from rewriting the buffer area for a specific retry operation Buffer area associated data packet empties. For this reason, the UE timer must be set to a value lower than the value of the Node B timer to initiate this operation in a timely manner, that is, for example, a retransmission data packet arriving at the Node B before Expiration of the time to be able to send.

If If the UE is in a Scheduled Mode, it must have the Capacity Request message (resource request message) to the Node B send. This capacity requirement can contain various attributes to the Node B at the meeting to support the scheduling decision. Such For example, parameters can be the priority the shipment, the amount of data in the buffer, and the time that to disposal stands until the package has to be sent. In another aspect considered the invention the UE the timer for generating the UE capacity request also for Setting the parameters sent in the capacity request.

It was explained that the timer corresponding to the worst case, assuming the continuous transmission of data packets from the UE and a immediate reuse of the same HARQ process and the same Sequence number through a new data packet, can be calculated. This is done to improperly combine the soft buffer values a package under all circumstances to avoid. On the other hand, such a case actually occurs only rarely and it is more likely that the soft-buffer values of Deleted time by time although there is still a slightly delayed retransmission can. In a further embodiment of the present invention, the timer to a higher value than in the described worst case Case are set. In this case, the UE must combine a faulty one in extreme situations that are close to the worst case, through appropriate processes, which is explained below be prevented.

In dependence From the feedback protocol, the UE knows the HARQ contexts of some or all Node Bs, that is, the status of the various processes or the ARQ window, the Timer, soft buffer need, and so on. If the ACK / NAK sent by all Node Bs, the UE may be due to a missing ACK / NAK know that Node B is most likely a specific one Missed package. If this was a first transfer (of a package), it is obvious that the soft buffer of Node B is not updating and therefore not with the buffers of the other Node Bs, that sent a feedback message is synchronized. Consequently, the UE can predict when an error situation is potentially can occur and avoid this case.

Since the error situation only occurs when a new packet is sent, the sender can use a different HARQ process and / or sequence number (or NDI) for the new packet, thus avoiding potentially erroneous combining. If the soft buffer space is limited, a HARQ process and / or sequence number (or NDI) value must be available that is not in the processing state or that has values stored for combining. When the entire soft buffer is in use, the corresponding retransmission operation may be delayed until an ACK for another packet is received in the soft buffer and this HARQ process and / or sequence number (NDI) can be reused. Also, the HARQ process may be delayed until the timer expires and restarts with the same packet. In general, it should be avoided that the UE must wait until the timer expires in the Node B to reduce latency. Nevertheless, there may be special cases depending on the data packet retransmission protocol design in which the expiration of the timer is desirable. Instead of waiting for the timer to expire in the Node B, the UE may itself initiate some or all of a soft buffer area in the Node B. This can be done, for example, by using a clear indicator, by artificially shifting the HARQ window forward, corresponding to emptying a portion of the entire buffer, or by simply exploiting some packets that are not acknowledged. If the UE is unable to transmit packets repeatedly, for example because the transmit power is insufficient, timely scheduling of retransmissions is not possible, and so forth, it may signal this to Node B to allow clearing of the soft buffer avoid. This can be done, for example, by a flag along with the other HARQ-related control information, such as the HARQ process number and the sequence number (or NDI). A special restart request message including the flag may be sent by the UE to direct a Node B to restart a particular current timer. Upon receipt of the message, the Node B halts the timer for that packet or preferably reboots it and maintains the soft buffer. Another possibility would be a kind of zero payload packet with certain control information, but with a transport block size of zero. This means that no real data is sent. Although this program consumes some resources, it can be more funky-efficient than emptying the soft buffer, which is nearly enough red may have accumulated for successful decoding.

As described above, in accordance with an embodiment of the present invention, it is desirable for a single Node B to send feedback messages to the UE to acknowledge a received data packet. For this reason, the selection of the Serving Node B may require further consideration to provide reliable feedback to the UE. Possible selection criteria with regard to radio link quality indicators for the selection of a serving node B are described in the parallel application "serving base station selection during soft handover", EP 1 507 422 A1 , explained.

The Fact that the for the synchronization of the soft buffer contents used timers is about to expire, may be considered deterioration of the uplink radio link conditions this particular Node B will be interpreted. Signaling this information for support the reselection of Serving Node B depends on the considered UTRAN architecture from. In the R99 / 4/5 architecture, the information may be from the current one Serving Node B are signaled to the RNC. In the advanced Architecture can however, the radio-related protocol units are located in the Node B + s be. It may be the current serving node B +, a new one Serving Node B + to select and the decision to signal him. Consequently, in In this case, the fact that the timer is in the current serving node B is about to expire, not signaled to other network elements become.

The parallel application describes the negotiation of the activation time for the selection of Serving Node B. One possible interaction with the present application would be to consider the status of the soft buffer synchronization timer before suggesting a new activation time. Depending on the architecture of the radio access network (RAN), the actual use, the transport technique and so on, there may be various delays at the Iub / Iur interfaces. Depending on these delays, it may be advantageous to use the invention described above or the soft buffer synchronization by signaling between the Node Bs of the Active Set described in the copending application EP 1 507 421 A1 described. With short signaling delays within the network (e.g., all Node Bs are part of the same or Radio Network Subsystem (RNS)), it may be advantageous to use a synchronization method as described in the co-pending application, while for longer delays the present invention is preferred can. Both methods can also be used in parallel or in dependence on each other. When the signaling has arrived, the timer is replaced or vice versa.

A another embodiment represents an alternative solution the above-described object of the present invention ready. If a packet is stored in the Node B and the Node B for one does not receive a retransmission, it does not know if it is retransmitting who may have received correctly from another Node B. or really nothing was sent in the uplink.

If there are no errors in the downlink feedback signaling, the UE knows its transmission status (HARQ context) of each Node B exactly. Even if some Node Bs do some retransmissions all the way this is the UE on the basis of the missing feedbacks known by these UEs.

In this embodiment can be an extra Flag that indicates whether the received data packet to be combined with previous broadcasts. If a package, which is still stored in the soft buffer, not in the meantime is assigned or not confirmed by any of the Node Bs, For example, a combine indicator may be set by the UE to one Node B indicate that the package can still be combined. Thereby A Node B is given the guarantee that the package is still combined can be. At a first transmission of a new data packet, the Combine Indicator may indicate that the received packet not with the previously received data packets combine and the Node B can have a buffer area which the process number signaled with the currently received data packet corresponds, delete. An advantageous combination with the deletion timer can also possible be. When the timer has expired and a packet with a Unset Combined Indicator becomes the package discarded.

If at a pending retransmission the channel conditions changed in the meantime It may also be advantageous to the transmission format of the package to change. This implies that no combining is possible. In this case can the deletion bit be set, though the same package with the same sequence number is sent. The recipient can clear the HARQ buffer, although a package with the same sequence number (or process number and NDI) is still cached. An advantage in comparison to the increase The sequence number is that there is no packet in the reorder buffer is missing.

at fast cell site selection (FCS) the resulting problems are similar those described in the preamble of this application. In contrast, there may be no soft handover for a UE, but one fast switching between different cells.

In a further embodiment of the present application based on the present invention described above principles also apply to HARQ soft buffer synchronization while fast cell site selection (FCS) become. Using the Fast Cell Site Selection, the UE always to a single cell, preferably to the cell with the best channel characteristics or lowest load (no soft-handoff transmission). Dependent on of the detailed protocol, the UE can within a certain time or switch each TTIs between cells. The cell change can independently performed by the UE or wholly or partially controlled by the network. On the same As in a soft handover, the soft buffer must be synchronized, before the next Shipment can arrive at the same Node B. For FCS, that can be for synchronization available Take into account a change from this cell and back to that cell. The same further embodiment can be applied if there are different coordination modes, one mode being characterized by the support of the HARQ can be while another one is not. When the UE is in the coordination mode HARQ can not support it while it is in it independent Mode, such functionality can support. One of the reasons for that can be that for the coordination mode more control information between the UE and must be signaled to the Node B. These could with the for control information required by the HARQ process. When back or forward is changed, the soft buffer can also be synchronized.

It is possible, that the RNC does not care about a mode change or a cell change who knows performed by the Node B and the mobile terminal. Once the RNC, the serves as a rearrangement unit, packets from a new Node B receives can he the previous or all inform other Node Bs in the Active Set of their soft buffers to empty. Alternatively, the Node B may know if a cell change is carried out and can inform the old Node B or the other Node Bs about it. The other Node Bs can empty their buffers accordingly. The Node B, which is the mode or The cell change is also known, the RNC in the Rel99 / 4/5 architecture or the current Serving Node B + in the evolved architecture (evolved architecture) about it to inform. The RNC or Serving Node B + can also use the inform other Node Bs of the Active Set about their buffers to empty accordingly. When the cell or mode change slow and not package for Package is carried out The soft buffer can be synchronized before a cell or Mode change back to the previous cell or mode.

As described above, may have the disadvantage that soft buffer values to be deleted, although retransmissions can still arrive. When very fast cell or mode change, for example, per TTI, There is a high probability that a repeated selection same cell or mode is common. In this case It may be advantageous to set the soft buffer values for one potential change back to the cell or mode for to keep a certain amount of time. This will be a combination allows retransmissions with previous broadcasts, already cached in the soft buffer. It can also happen that the communication terminal or the mobile terminal after the change sends data to a new cell or mode. In this case you can decide back to the previous cell or switch to the previous mode and send with it Status of the associated buffer area continue. The time until an emptying is performed can again be defined as a threshold time period, wherein at least one of the base stations and the communication terminal the time which has elapsed since the storage of the data packet in the associated buffer area is, monitor can. The threshold period may be similar to that described above Way possibly considering a additional Cell or mode change time calculated become. It may again be a conflict of interest between the minimum length of Threshold and profit by soft combining. By the method described or by combining with other methods can avoid erroneous combining. The described threshold time for FCS or a mode change can equal the threshold time span for the Soft handover, that is, of the plurality of base stations, or be different from this. If the value is different, he can also at least one of the respective base stations by radio network control signaling and to the communication terminal by radio resource signaling to a similar one as described above Be signaled way.

Finally, it is pointed out that the present invention described above can be used for various types of RAN architectures. For example, the present invention is directed to the in 2 illustrated UMTS R99 / 4/5 UTRAN architecture as well as the in 4 illustrated evolved UTRAN architecture applicable.

Claims (36)

Method for avoiding soft buffer corruption in a data retransmission protocol in a mobile communication system, this is a communication terminal and a plurality of base stations, wherein the communication terminal during a Soft handover communicates with the multitude of base stations and the method comprises: Receiving data from the communication terminal at least one of the base stations, Save the received Data in an associated Soft buffer of at least one of the base stations, Decode the received data at the at least one of the base stations, and Benefits of saving the data in the associated soft buffer elapsed time to clear the soft buffer. Method for receiving data according to claim 1, where data is in the associated soft buffer stored in memory in order to be combined with repeatedly sent data, if the received data is not successfully decoded. Method for receiving data according to claim 1 or 2, the further rooms of the soft buffer based on the elapsed time. The method of claim 3, wherein when the received Data is successfully decoded, the soft buffer is evacuated. Method according to claims 3 or 4, wherein the soft buffer vacated when the elapsed time equals a threshold period is or longer as this one. The method of claim 5, wherein the threshold period is defined as a period after the retransmission data in the base station can no longer be expected. The method of claims 1-6, further comprising: Receive of retransmission data from the communication terminal at least one of the base stations, Save the received Retransmission data in an associated soft buffer, Decode the received retransmission data, and if the received Retransmission data will not be decoded successfully, again Start the elapsed time. The method of any of claims 1-6, further comprising: Receive of retransmission data from the communication terminal at least one of the base stations, Save the received Retransmission data in an associated soft buffer, Decode the received retransmission data, and if the received Send Repeat Data is successfully decoded, pausing the elapsed time. The method of any of claims 1-8, further comprising: Combine from retransmission data with previously received data to combined To generate data, and Decode the combined data. The method of any of claims 1-9, further comprising the elapsed time equals or exceeds a threshold period this includes stopping the elapsed time. The method of claim 5 or 10, wherein the threshold period of configurable duration. The method of claim 11, further comprising: Signal the duration of the threshold period to at least one of the base stations by radio network control signaling from a control unit in the Mobile communication network. The method of claim 12, wherein the duration of the Threshold period of the at least one of the base stations in one Information element of an NBAP message is signaled. The method of any one of claims 11 to 13, further comprising includes: Signaling the duration of the threshold period to the Communication terminal through Radio resource control signaling from a control unit in the Mobile communication network. The method of claim 14, wherein the duration of the threshold period is communicated to the communication terminal in an information element of at least one radio bearer setup message, Radio Be arer reconfiguration message, a Radio Resource Control Connection Setup message, a Transport Channel Reconfiguration message, a Cell Update message, and a Handover Command message. The method of any one of claims 1-15, further comprising: Send a message from at least one of the base stations to the communication terminal indicating whether the at least one of the base stations receive the received data successfully decoded. The method of any of claims 1-16, further comprising: Hand off the received data to a control unit of the mobile communication system by at least one of the base stations receiving the received data successfully decoded. The method of any of claims 1-17, further comprising: Receive a capacity request message from the communication terminal at least one of the base stations, the additional transmission capacity for retransmission data requesting. The method of claim 18, wherein the capacity request message at least one transmission priority of data to be transmitted by the communication terminal, the size of data in a transmission buffer of the communication terminal, the duration of the elapsed Time, which includes identification of data or the channel, requested for the capacity becomes. The method of claim 18 or 19, further includes: Sending a capacity grant message from the at least one one of the base stations to the communication terminal, wherein the capacity grant message a transmission capacity assigned to the communication terminal for data transmission becomes. The method of any one of claims 1-20, wherein the data is over a dedicated channel are received. The method of any of claims 1-21, further comprising: Receive a restart request message from the communication terminal at least one of the base stations, wherein the restart request message Data indicates, for to restart the elapsed time. The method of claim 22, wherein the restart request message Control information and no or empty payload includes. The method of claim 5 or 10, wherein the data retransmission protocol is a window-based data retransmission protocol and that A method of further calculating the threshold period based on includes time for sending all data within a window of the data retransmission protocol is required. The method of claim 5 or 10, further includes: Calculate the threshold period based on the time interval between the receipt of output data and the reception of retransmission data. The method of claim 5 or 10, further includes: Calculate the duration of the threshold period based on the size of the soft buffer, the maximum Number of retransmissions in the data retransmission protocol, the processing time of the communication terminal for a feedback message, the processing time the corresponding base station and a transmission time interval. Base station in a mobile communication system, the one communication terminal and one Variety of base stations, wherein the base station a Means for implementing the method according to any one of claims 1-26. Base station for avoiding soft buffer corruption in a data retransmission protocol in a mobile communication system, this is a communication terminal and a plurality of base stations, wherein the communication terminal during a Soft handover communicates with the multitude of base stations and the Base station includes: a receiving section that receives data from the communication terminal receives one Soft buffer that stores the received data, and one Decoding section that decodes the received data, in which the base station can be operated so that it has been saving the data in the associated Soft buffer uses elapsed time to clear the soft buffer. The base station of claim 28, wherein the base station can be operated so that they receive the received data in the associated Stores soft buffer to combine with re-sent data, if the received data is not successfully decoded. A base station according to claim 28 or 29, wherein the Base station can be operated so that they are the soft buffer admits when the received data is successfully decoded. A base station according to any of claims 28-30, wherein the base station can be operated to provide the soft buffer admits if the elapsed time is equal to a threshold period or longer as this one. The base station of claim 31, wherein the threshold period is defined as a period after the retransmission data in the base station can no longer be expected. The base station according to claim 28-32, wherein the receiving section can be operated so that it retransmission data from the Receive communication terminal, the Soft buffer can be operated so that it receives the received retransmission data stores, and the decoding section can be operated so that it decodes the received retransmission data, the Base station can be operated so that when the received Send Repeat Data is not successfully decoded, the elapsed Time starts again. A base station according to any of claims 28-32, wherein the receiving section is operable to receive retransmission data from the communication terminal receives the soft buffer can be operated so that it receives the received Retransmission data stores, and the decoding section so can be operated to receive the received retransmission data decoded, wherein the base station can be operated so that it if the received retransmission data is successfully decoded that lasts for elapsed time. A base station according to any one of claims 28-34, wherein the base station can be operated to send retransmission data combined with the previously received data to combined data to generate, and the decoding section can be operated so that he decodes the combined data. A base station according to any of claims 28-35, wherein the base station can be operated so that it passes the elapsed time Time stops, if the elapsed time is equal to a threshold period or longer as this one.